Implications of Progesterone Receptor Status for the Biology and Prognosis of Breast Cancers

日大医誌 75 (1): 10 15 (2016) 10 Original Article Implications of Progesterone Receptor Status for the Biology and Prognosis of Breast Cancers Naotaka Uchida 1), Yasuki Matsui 1), Takeshi Notsu 1) and Manabu Yoshida 2) 1) Department of Breast, Endocrine, Vascular and Thoracic Surgery, Matsue City Hospital 2) Department of Clinical Pathology, Matsue City Hospital Background: It remains unknown whether knowledge of the status of progesterone receptor (PR) expression is useful for distinguishing between luminal A and B breast cancer subtypes and for providing an accurate prognosis for patients with estrogen receptor-positive (ER+) and human epidermal growth factor receptor 2-negative (HER2 ) breast cancer. We aimed to assess the role of PR status in determining the biology and prognosis of early ER+ and HER2 breast cancer. Methods: This was a retrospective study of 105 patients with ER+ and HER2 breast cancer who underwent surgery between 2005 and 2013. Relapse-free survival (RFS), distant-metastasis-free survival (DFS) and overall survival (OS) were evaluated. The prognostic value of PR was evaluated using a Cox regression model. Results: Apart from the menstruation status, there were no significant differences between the PR+ and PR patients. PR patients exhibited a tendency towards a higher nuclear grade and Ki-67 labeling index. In particular, postmenopausal patients with PR expression status < 10% exhibited a significantly higher nuclear grade and tendency towards a higher Ki-67 labeling index. After a median follow-up of about 5 years, the 5-year RFS, DFS and OS rates of the PR+ and PR patients were not significantly different. PR was not a significant prognostic factor by multivariate analysis. Conclusion: PR status does not have prognostic value for patients with early ER+ and HER2 breast cancer who received adequate therapies. However, it may be reasonable to use PR status for distinguishing between luminal A and B breast cancer subtypes. Key words: progesterone receptor, breast cancer, prognosis (J. Nihon Univ. Med. Ass., 2016; 75 (1): 10 15) Introduction For patients with breast cancer, assessment of the PR status of their tumor is considered important for differentiating between luminal A and B subtypes and for predicting responsiveness to endocrine therapy. Based on a study by Prat et al., the 2013 St. Gallen guidelines confirmed that PR has added value for distinguishing between luminal A and B breast cancer subtypes that are determined by immunohistochemical (IHC) analysis, as described in the 2011 St. Gallen guidelines 1 3). Tumors with decreased or no expression of PR have poorer response to selective ER modulators such as tamoxifen (TAM) 4) and better response to aromatase inhibitors (AI) 5). However, PR status has also been reported to be not significant for predicting response to endocrine therapy. The Early breast cancer trialists collaborative group (EBCTCG) overview of all randomized trials of adjuvant TAM therapy for early breast cancer reported that PR status did not predict response to TAM 6). In addition the AI exemestane was reported to be equally effective for ER+/PR+ and ER+/PR breast cancer 7). Thus, the role of PR status in determining the prognostic and biologic characteristics of breast cancers has been controversial. We aimed to assess the role of PR status in determining the biology and prognosis of breast cancers, by retrospectively analyzing patients who underwent surgery and adjuvant therapy for early ER+ and HER2 breast cancer. Patients and Methods Patients A total of 105 patients with invasive breast cancers who underwent surgery in our hospital between September 2005 and September 2013 were investigated retrospectively. Men with breast cancer and women with in situ carcinoma, bilateral breast carcinoma, tumors less than 0.5 cm, or who underwent neoadjuvant chemotherapy or chemotherapy for another disease were excluded from the study. Informed consent was obtained from every patient. Decisions regarding treatment were made by the treating physician on the basis of patient preference, type and stage of disease and recurrence risk factors. Patients were treated with breast-conserving surgery or mastectomy. Most patients who underwent breast- Received: June 19, 2015 Accepted: September 30, 2015

75 (1) 2016 Role of Progesterone Receptor in Breast Cancer 11 conserving surgery received radiotherapy to the whole breast after surgery. Most patients received postoperative endocrine therapy. Some patients received post-operative chemotherapy. Histopathological and immunohistochemical evaluations All histopathological and immunohistopathological diagnoses were determined by pathologists at our facility or at another laboratory facility within our partnership. Surgical specimens were embedded in paraffin, sectioned and stained with hematoxylin and eosin. ER positivity, PR positivity and Ki-67 labeling index were determined by immunohistochemical (IHC) analysis. Tumors were considered ER+ if 1% of tumors cells were stained positive for ER expression. Tumors were considered PR+ if >20% of tumors cells were stained positive for PR expression. HER2 expression status was determined by IHC analysis, with scores of 0, 1+, 2+ indicating the following: no cells with membrane staining and <10% of cells with membrane staining, 10% of cells with slight membrane staining, 10% of cells with low or medium membrane staining, respectively. Tumors that were 2+ by IHC analysis were also examined by fluorescence in situ hybridization, with an amplification ratio < 2.0 indicating negative status. Statistics Relapse-free survival (RFS) was defined as the period from the date of operation to the date of the first confirmation of relapse (i.e., local relapse or metastasis) or death from any cause, whichever came first. Distant-metastasis-free survival (DFS) was defined as the period from the date of operation to the date of the first confirmation of distant metastasis or death from any cause, whichever came first. Overall survival (OS) was defined as the period from the date of operation to the date of death from any cause. RFS, DFS and OS were estimated using the Kaplan- Meier method and compared using the log-rank test. The Mann-Whitney U test was used for comparisons of continuous outcomes, while the chi-square or Fishers exact test was used for comparisons of categorical variables. Risk factors affecting prog no sis were assessed by multivariate analysis using Cox proportional-hazards model. Differences were considered significant for P < 0.05. Results Patient characteristics Table 1 shows the characteristics of 105 breast cancer patients. There were no significant differences in age, body mass index, surgical procedure, tumor size, number of metastatic axillary lymph nodes, proportion of patients with ER+ tumors, nuclear grade, Ki-67 labeling index, proportion of patients undergoing radiation therapy after-breast conserving surgery, proportion of patients undergoing postoperative chemotherapy, and proportion of patients undergoing post-operative endocrine therapy between the PR+ and PR patients. Among the PR+ and PR patients, menstruation status was significantly different. The proportion of postmenopausal patients was higher among PR than PR+ patients. PR patients had a tendency to higher nuclear grade and Ki-67 labeling index, although the differences were not significant. Assessment of the Ki-67 labeling index was initiated in 2009, and the number of patients with Ki- 67 labeling index values was limited. Chemotherapy was significantly more frequently administered to patients who were younger, with larger tumors, more metastatic lymph nodes, higher Ki-67 labeling indices, and tumors with higher nuclear grade. Among patients who did or did not undergo chemotherapy, there were no statistical differences in the proportions of those with ER+ and PR+ tumors. Table 2 shows the clinicopathological features between premenopausal and postmenopausal patients. The proportion of ER was significantly higher among postmenopausal than premenopausal patients, while the proportion of PR was significantly lower among postmenopausal than premenopausal patients. Postmenopausal patients had a tendency to higher nuclear grade and Ki-67 labeling index, although the differences were not significant. Table 3 shows the nuclear grade and Ki-67 labeling index in postmenopausal patients. When 10% was used for the cut-point of PR, lower PR expression was significantly associated with higher nuclear grade and had a tendency to higher Ki-67 labeling index. Relapse-free, distant-metastasies-free, and overall survival The RFS rates were estimated for PR+ and PR patients. The median follow-up period was 61 months. The estimated 5-year RFS rates of PR+ and PR patients were 93.6% and 90.1%, respectively (P = 0.742, Fig. 1). There were 10 patients who developed relapse, including 2 with local and regional lymph node recurrences and 8 with distant metastases. The median time to relapse was 54 months (range 9 77 months). The DFS rates were estimated for PR+ and PR patients. The median follow-up period was 62 months. The estimated 5-year DFS rates of PR+ and PR patients were 93.5% and 100%, respectively (P = 0.094, Fig. 2). There were 8 PR+ and no PR patients with distant metastasis. The median time to distant metastasis was 56 months (range 21 77 months). There were 4 patients with pulmonary metastasis, 3 with bone metastasis, 2 with ovarian metastasis, and 1

12 Naotaka Uchida et al. Characteristic Table 1 Patients characteristics PR+ group n = 79 PR group n = 26 P-value Age-year Median 55 59 0.262 Range 32 95 36 81 Menstruation status-no. (%) Premenopause 31 (39) 4 (15) 0.030 Postmenopause 48 (61) 22 (85) BMI Median 22.4 22.9 0.803 Range 15.7 34.4 16.6 29.8 Surgical procedure-no. (%) mastectomy 14 (18) 2 (8) 0.357 breast-conserving surgery 65 (82) 24 (92) Tumor size-cm Median 1.5 1.5 0.797 Range 0.5 4.5 0.6 3.5 Lymph node metastasis number Median 0 0 0.479 Range 0 4 0 1 ER positivity proportion-% Median 90 90 0.429 Range 60 100 1 100 Nuclear grade-no. (%) Grade 1 48 (61) 14 (54) 0.534 Grade 2 or 3 31 (39) 12 (46) Ki67-% Median 17 34 0.799 Range 1 85 5 76 not examined 55 16 Radiation therapy after breast conserving surgery-no. (%) 60 (92) 23 (96) 0.556 Post-operative chemotherapy-no. (%) 29 (37) 13 (50) 0.230 Post-operative endocrine therapy-no. (%) 76 (96) 25 (96) 0.991 Table 2 Clinicopathological features between premenopausal and postmenopausal patients Features Premenopuase n = 35 Postmenopause n = 70 P-value ER positivity proportion-% Median 90 100 0.003 Range 60 100 1 100 PR positivity proportion-% Median 80 60 0.014 Range 1 100 0 100 Nuclear grade-no. (%) Grade 1 22 (63) 40 (57) 0.574 Grade 2 or 3 13 (37) 30 (43) Ki67-% Median 17.1 17.8 0.732 Average 18.7 24.3 Range 1 58 1 84 not examined 23 48

75 (1) 2016 Role of Progesterone Receptor in Breast Cancer 13 Table 3 Nuclear grade and Ki-67 labeling index in postmenopausal patients PR 10% n = 56 PR < 10% n = 14 P-value Nuclear grade-no. (%) Median 1 2 0.011 Range 1 3 1 3 Ki-67 labeling index-% Median 17 45 0.130 Range 1 84 4 76 not examined 40 8 with hepatic metastasis. The OS rates were estimated for PR+ and PR patients. The median follow-up period was 63 months. The estimated 5-year OS rates of PR+ and PR patients were 96.8% and 100%, respectively (P = 0.221, Fig. 3). There were 4 PR+ patients and no PR patients who died. The median time to death was 66 months (range 23 92 months). Two patients died of breast-cancer-related causes, 1 died of senility-related causes, and 1 died of acute pyelitis. Figure 1 % 100 Relapse free survival 80 60 40 20 p=0.742 0 0 1 2 3 4 5 6 7 8 9 10 number at risk 79 79 73 62 54 45 28 20 8 1 26 26 25 23 22 13 10 7 6 1 years Fig. 1 Relapse-free survival. There is no significant difference between PR+ and PR patients. Figure 2 % 100 Distant metastases free survival 80 60 40 20 p=0.094 number at risk 0 0 1 2 3 4 5 6 7 8 9 10 79 26 79 75 62 55 45 28 20 8 1 26 26 24 23 15 11 8 6 1 Fig. 2 Distant-metastasis-free survival. There is no significant difference between PR+ and PR patients. years

14 Naotaka Uchida et al. Figure 3 % 100 Overall survival 80 60 40 20 p=0.221 0 0 1 2 3 4 5 6 7 8 9 10 number at risk 79 79 74 64 57 48 34 26 12 3 26 26 26 24 23 15 11 8 6 1 years Fig. 3 Overall survival. There is no significant difference between PR+ and PR patients. Table 4 Relapse free survival, distant metastases free survival and overall survival (Multivariate analysis) Relapse free survival Multivariate analysis Hazard ratio 95% Confidence interval P-value Menstruation status (postmenopause vs. premenopuase) 0.93 0.59 1.45 0.763 Tumor size (>1 cm vs. 1 cm) 0.78 0.48 1.24 0.301 Lymph node involvement (positive vs. negative) 0.88 0.52 1.50 0.657 PR positivity (positive vs. negative) 1.04 0.64 1.67 0.872 Distant metastases free survival Multivariate analysis Hazard ratio 95% Confidence interval P-value Menstruation status (postmenopause vs. premenopuase) 0.93 0.60 1.45 0.768 Tumor size (>1 cm vs. 1 cm) 0.82 0.52 1.31 0.423 Lymph node involvement (positive vs. negative) 0.87 0.51 1.47 0.616 PR positivity (positive vs. negative) 1.05 0.66 1.67 0.816 Overall survival Multivariate analysis Hazard ratio 95% Confidence interval P-value Menstruation status (postmenopause vs. premenopuase) 0.98 0.63 1.53 0.958 Tumor size (>1 cm vs. 1 cm) 0.81 0.51 1.28 0.370 Lymph node involvement (positive vs. negative) 0.66 0.39 1.09 0.108 PR positivity (positive vs. negative) 0.93 0.59 1.47 0.776 Risk factors associated with relapse, distant metastasis, and death Risk factors for relapse, distant metastases and death were investigated by multivariate analysis. Independent variables included menstruation status, tumor size, lymph node involvement and PR positivity. No significant independent risk factors were identified for occurrence of relapse, distant metastasis, or death (Table 4). Discussion The predictive significance of the PR status of a breast malignancy has been controversial with regard to response to endocrine and/or chemotherapy. In a study that found that PR status was independent predictive factor for identifying breast cancer patients who would benefit from adjuvant endocrine therapy, the relative risk of recurrence was reported to be reduced by 53% for ER+/PR+ patients and 25% for ER+/PR patients among those who received

75 (1) 2016 Role of Progesterone Receptor in Breast Cancer 15 adjuvant endocrine therapy 4). In that study, about 50% of patients had positive axillary lymph nodes or tumors larger than 2 cm. Eighty-six percent of study patients were older than 50 years, suggesting that postmenopausal women were the main population of the study. However, they were treated using TAM, but not AI. Results of anastrozole, TAM alone or in combination (ATAC) trial suggested that patients with ER+/PR tumors were best treated by estrogen withdrawal using AI 5). In our study, 23% of patients had positive axillary lymph nodes, and 32% of patients had tumors larger than 2 cm. Sixty-six percent of study patients were postmenopausal women and they received AI, which are superior to TAM for the treatment of postmenopausal breast cancer 8). The earlier stage of prognostic factors and higher recurrent inhibition effect by AI in our study could result in no significant difference of RFS, DFS, and OS between PR+ and PR patients. The significance of PR expression status for distinguishing between luminal A and B breast cancer subtypes has also been a question. In two independent cohorts data sets, patients with IHC-luminal A tumors having low positive PR+ tumor cells ( 20%) showed significantly poorer survival compared with tumors with >20% of PR+ tumor cells 3,9,10). Especially, the low PR expression has had a high correlation with poor prognosis in postmenopausal patients 11). PRlow or negative luminal B-like breast cancer showed significantly higher histological grade than that of PR-high luminal B-like breast cancer 12). In our study, nuclear grade and Ki-67 labeling index were not significant different between PR+ group and PR group, but showed tendency to higher nuclear grade and Ki-67 labeling index in PR group. In postmenopausal women, PR expression was inversely correlated with nuclear grade significantly. Ki-67 labeling index also had similar tendency. PR is a marker of a functional ER, and the expression of PR is induced by a sufficient level of estrogen. The lower estrogen milieu in postmenopausal women leads to lower ER activity, and the prognostic effect of PR could be more important than that of the premenopausal women. These results suggest that low and negative PR expression levels reflects the worse property of breast cancer and the additional assessment of PR status for differentiating between luminal A and B subtypes is reasonable. The statistically insignificant results of Ki-67 labeling index might be accounted for by the inadequate power of our study. In conclusion, PR status does not have prognostic value for patients with low-risk ER+ and HER2 breast cancer who received adequate therapy. However, it may be reasonable to use PR status for distinguishing between luminal A and B subtypes. Conflict of Interest The authors declare that they have no conflict of interest. References 1) Goldhirsch A, Wood WC, Coates AS, et al. Strategies for subtypes dealing with the diversity of breast cancer: highlights of the St. Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2011. Ann Oncol 2011; 22: 1736 1747. 2) Goldhirsch A, Winer EP, Coates AS, et al. Personalizing the treatment of women with early breast cancer: highlights of the St Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2013. Ann Oncol 2013; 24: 2206 2223. 3) Prat A, Cheang MC, Martín M, et al. Prognostic significance of pro ges terone receptor-positive tumor cells within immuno histo chemically defined luminal A breast cancer. J Clin Oncol 2013; 31: 203 209. 4) Bardou VJ, Arpino G, Elledge RM, et al. Progesterone receptor status significantly improves outcome prediction over estrogen receptor status alone for adjuvant endocrine therapy in two large breast cancer databases. J Clin Oncol 2003; 21: 1973 1979. 5) Dowsett M, Cuzick J, Wale C, et al. Retrospective analysis of time to recurrence in the ATAC trial according to hormone receptor status: an hypothesis-generating study. J Clin Oncol 2005; 23: 7512 7517. 6) Early Breast Cancer Trialists Collaborative Group. Tamoxifen for early breast cancer: an overview of the randomised trials. Lancet 1998; 351: 1451 1467. 7) Coombes RC, Hall E, Gibson LJ, et al. A randomized trial of exemestane after two to three years of tamoxifen therapy in postmenopausal women with primary breast cancer. N Engl J Med 2004; 350: 1081 1092. 8) Iwase H. Current topics and perspectives on the use of aromatase inhibitors in the treatment of breast cancer. Breast Cancer 2008; 15: 278 290. 9) Martín M, Rodríguez-Lescure A, Ruiz A, et al. Randomized phase 3 trial of fluorouracil, epirubicin, and cyclophosphamide alone or followed by Paclitaxel for early breast cancer. J Natl Cancer Inst 2008; 100: 805 814. 10) Cheang MC, Chia SK, Voduc D, et al. Ki67 index, HER2 status, and prognosis of patients with luminal B breast cancer. J Natl Cancer Inst 2009; 101: 736 750. 11) Nishimukai A, Yagi T, Yanai A, et al. High Ki-67 Expression and Low Progesterone Receptor Expression Could Independently Lead to a Worse Prognosis for Postmenopausal Patients with Estrogen Receptor- Positive and HER2-Negative Breast Cancer. Clin Breast Cancer 2015; 15 (3): 204 211. 12) Sato K, Miyashita M, Ishida T, et al. Prognostic significance of the proges terone receptor status in Ki67-high and -low Luminal B-like HER2-negative breast cancers. Breast Cancer. 2014 Nov 8. DOI 10.1007/s12282-014- 0575-6